Design of English Translation Model Based on Recurrent Neural Network

English is a language that is spoken by almost every person on this planet. It has become the most important language for all the people of this world. English is not only used in business but also in personal life. People from different countries use English as their first language and they want to communicate with other people who are using other languages. This paper aims to discuss about how machine learning can be used to design an effective model for translation between two languages like English and Russian, which will help us to improve the services provided by machines as well as human translators.The create a model that can both predict and generate the meaning of English sentences. This is done using a recursive neural network (RNN). RNN works on sequential input, which encodes information according to the previous example. The proposed RNN model is based on three different layers: input layer (layer 1), which uses context and word embedding as features; The middle layer (layer 2) containing LSTM units is used to predict new words according to the context of new words and previous contexts; And the output layer (layer 3), which considers the meaning of words and their context information. Based on these inputs, RNN will output the next word or sentence, even the paragraph that should be used to predict the meaning. The model was created using the Keras library, which makes it easy to use recurrent neural networks.